摘要
针对宽脉冲在磁声电成像(magneto-acousto-electrical tomography,MAET)方法中存在纵向分辨率低的问题,对基于线性调频(linearly frequency-modulated,LFM)宽脉冲的MAET方法提出一种重建算法,通过对信号与图像的处理恢复隐藏在脉冲压缩信号中的分界面信息。首先去除接收信号经过脉冲压缩滤波器滤波后的信号中由串扰信号引起的部分噪声,之后利用图像分割思想,对单通道信号中的MAET信号进行提取,根据提取信息重新计算叠加信号的幅值与位置,最后重绘成像结果。该算法相较于传统的B扫模式成像能有效地解决近距离磁声电信号的脉冲压缩信号主瓣部分相互掩盖的问题,同时在保持激励信号带宽与系统电路成本不变的情况下,有效提高基于LFM的MAET方法的纵向分辨率。在MATLAB中进行仿真计算,搭建MAET系统,对仿体进行成像实验,对比B扫模式成像与该算法成像的性能。结果表明,该方法能够有效地区分基于LFM的MAET方法所获得的叠加分界面信号,并且重建准确率高。
To address the issue of low longitudinal resolution in wide pulse magneto-acousto-electrical tomography(MAET),a reconstruction algorithm based on linearly frequency-modulated(LFM)pulse is proposed in this paper.The interface information hidden in pulse compression signal was recovered by signal and image processing.First,part of the noise caused by the crosstalk signal in the received signal filtered by the pulse compression filter was removed.Subsequently,MAET signal in the single-channel signal was extracted using the idea of image segmentation,and the amplitude and position of the superimposed signal were recalculated according to the extracted information.Finally,the imaging results were redrawn.Compared with the traditional B-scan mode imaging method,this algorithm can effectively solve the problem of mutual cover-up of the main lobe of the pulse compression signals in the adjacent MAET signal,and can significantly enhance the longitudinal resolution of the MAET method based on LFM signal while keeping the bandwidth of the excitation signal and the cost of the system circuit unchanged.The algorithm was initially simulated in MATLAB,and then an MAET system was built to conduct imaging experiments on the sample,and the performance between the B-scan mode imaging and the algorithm was compared.Results show that this method can effectively distinguish the superimposed interface signals obtained by MAET method based on LFM with high reconstruction accuracy.
作者
孙直申
成智卓
SUN Zhishen;CHENG Zhizhuo(Faculty of Information Technology,Beijing University of Technology,Beijing 100124,China)
出处
《北京工业大学学报》
2025年第10期1179-1189,共11页
Journal of Beijing University of Technology
基金
国家自然科学基金资助项目(62001013)
北京市朝阳区博士后工作经费资助项目(2020ZZ-36)。
作者简介
孙直申(1985-),男,讲师,主要从事磁声电成像、弹性成像方面的研究,E-mail:zhishen.sun@bjut.edu.cn。
